Introduction to Base Graphics in R

Scatter Plot

# Generate sample data
set.seed(123); x <- rnorm(50)+1:50; y <- rnorm(50)+1:50

# Basic scatter plot
plot(x, y,
     main = "Scatter Plot",
     xlab = "X-axis",
     ylab = "Y-axis",
     pch = 19,       # solid circle
     col = "blue")

Line Plot

x <- 0:10
y <- x^2

plot(x, y,
     type = "l",
     main = "Line Plot",
     xlab = "X-axis",
     ylab = "Y-axis",
     col = "red")

Plot options

par(mfrow = c(2, 2))
plot(x, y, type = "l", main = "type = 'l'")
plot(x, y, type = "p", main = "type = 'p'")
plot(x, y, type = "b", main = "type = 'b'")
plot(x, type = "h", main = "type = 'h'")

Adding more data to the plot

ToothGrowth data set contains the results from an experiment to compare the effect of two supplements on tooth growth in 60 guinea pigs. A data frame with 60 observations on 3 variables.

  • [,1] len numeric Tooth length
  • [,2] supp factor Supplement type (VC or OJ).
  • [,3] dose numeric Dose in milligrams/day
len supp dose
4.2 VC 0.5
11.5 VC 0.5
7.3 VC 0.5
5.8 VC 0.5
6.4 VC 0.5
10.0 VC 0.5

Adding more data to the plot - simple plot

plot(ToothGrowth$dose, ToothGrowth$len,
     main = "Tooth Growth",
     xlab = "Dose (mg/day)",
     ylab = "Tooth length",
     pch = 19,       # solid circle
     col = "blue")

Adding more data to the plot - by supplement type

tg_vc <- ToothGrowth[ToothGrowth$supp == "VC", ]
tg_oj <- ToothGrowth[ToothGrowth$supp == "OJ", ]
plot(tg_vc$dose, tg_vc$len,
     main = "Tooth Growth by Supplement Type",
     xlab = "Dose (mg/day)",
     ylab = "Tooth length",
     pch = 19,       # solid circle
     col = "blue")
points(tg_oj$dose, tg_oj$len, pch = 19, col = "red")
legend("bottomright", legend = c("VC", "OJ"), col = c("blue", "red"), pch = 19)

plot function vs. points function

  • plot function creates a new plot
  • points function adds points to an existing plot
  • lines function adds lines to an existing plot
    • lines function is like points(type = "l")

Plotting categorical data

Here len ~ supp is a formula that tells R to plot len on the y-axis and supp on the x-axis.

boxplot(len ~ supp, data = ToothGrowth,
        main = "Tooth Growth by Supplement Type",
        xlab = "Supplement Type",
        ylab = "Tooth length"
)

Plotting categorical data

Input for boxplot function can be a formula, a matrix, data frame, or a list.

Categorical data - stripchart function

stripchart(len ~ supp, data = ToothGrowth,
           main = "Tooth Growth by Supplement Type",
           xlab = "Supplement Type",
           ylab = "Tooth length",
           col = "blue",
           vertical = TRUE
)

Stripchart is a one-dimensional scatter plot that represents the distribution of a continuous variable for different levels of a categorical variable. It can be used for similar data like boxplot.

Categorical data - stripchart function

What is points overlap? We can use jitter to spread the points out.

stripchart(len ~ supp, data = ToothGrowth,
           main = "Tooth Growth by Supplement Type",
           xlab = "Supplement Type",
           ylab = "Tooth length",
           col = "blue",
           method = "jitter",
           vertical = TRUE
)

Bar plot vs Box plot

  • Use boxplots when you need to display the full distribution, including medians, quartiles, and outliers
  • Boxplots are ideal for comparing the spread and distribution across multiple groups
  • Bar charts can mask the underlying data distribution; opt for boxplots when the detailed spread and potential outliers are important
  • Use Bar charts when you want to show a distribution of data points or perform a comparison of metric values across different subgroups of your data

Bar plot example

counts <- table(mtcars$gear) # frequency table
par(mfrow = c(1,2))
barplot(counts,
        main = "Bar Chart of Gears",
        xlab = "Number of Gears",
        ylab = "Frequency",
        col = "lightgreen")
# multiple bar plots
counts <- table(mtcars$am, mtcars$gear)
barplot(counts,
        main = "Bar Chart of Gears by Transmission",
        xlab = "Number of Gears",
        ylab = "Frequency",
        col = c("lightblue", "lightgreen"))

Histogram

Function hist takes a vector of values and computes a histogram.

x2 <- rnbinom(100, mu = 10, size = 10)
hist(x2,
     main = "Histogram of Negative Binomial Distribution",
)

Histogram

In default settings, hist calculates the number of bins automatically. You can specify the number of bins using the breaks argument.

par(mfrow=c(1,2))
hist(x2,
     main = "breaks = 20",
     breaks = 20
)
hist(x2,
     main = "breaks = 1:40",
     breaks = 0:40
)

Customizing plots

par(mfrow = c(2,3))
plot(x,y, main='default plot')
plot(x,y, col="#FF0000", main ="col set point color")
plot(1:20, col=rainbow(20), main ='col is a vector os colors')
plot(1:20, pch=1:20, main = "pch set point shape")
plot(x,y, cex = 3, main = "cex set point size")
plot(x,y, main = "plot with lines and grid")
abline(0, 2) # line with intercept 0 and slope 1
grid()

Customizing plots

par(mfrow = c(2,3))
plot(x,y, main='default plot', type='b')
plot(x,y, xlim=c(0, 30), ylim=c(0, 20), type='b', main = "setting limits of x and y axes")
plot(x,y, log='y', type='b', main = "log scale on y-axis")
plot(x,y, log='x', type='b', main = "log scale on x-axis")
plot(x,y, log='xy', type='b', main = "log scale on both axes")

Adding legends and annotations

plot(x, y, pch = 19, col = "blue")
points(x + 0.5, y + 0.5, col = "red", pch = 17)
# Add a legend to differentiate groups
legend("topleft",
       legend = c("Group 1", "Group 2"),
       col = c("blue", "red"),
       pch = c(19, 17))
# Add a mathematical annotation using expression()
mtext(expression(alpha + beta == gamma), side = 3, cex =2)

Combining multiple plots

Regular grid layout using par function

par(mfrow = c(2, 3))  # vector specifying the number of rows and columns
plot(x, y, main = "plot 1", col = "1")
plot(x, y, main = "plot 2", col = "2")
plot(x, y, main = "plot 3", col = "3")
plot(x, y, main = "plot 4", col = "4")
plot(x, y, main = "plot 5", col = "5")
plot(x, y, main = "plot 6", col = "6")

Combining multiple plots

Custom grid layout using layout function

layout(matrix(c(1, 2, 4, 3), nrow=2), width = c(2,1), height = c(1,2))
layout.show(4)

Combining multiple plots

Custom grid layout using layout function

layout(matrix(c(1, 2, 4, 3), nrow=2), width = c(2,1), height = c(1,2))
hist(x2, main = "1st plot")
plot(x, y, main = "2nd plot")
hist(rnorm(100), main = "3rd plot")

Exporting plots

  • png for bitmap graphics
png("outputs/plot1.png", width = 800, height = 600) # by defaults units are pixels
# whatever is plotted between `png` call and `dev.off` will be saved in the file
plot(x, y, main = "Scatter Plot", col = "blue")
dev.off()
png 
  2
  • Alternative output functions for bitmap graphics: jpeg, bmp, tiff
  • The output always include single plot. If you want to save multiple plots, use par or layout functions.

Exporting plots

  • pdf for vector graphics - also suitable for multi-page documents
  • svg for vector graphics Vector graphics is suitable if you want to do additional editing in a vector graphics editor like Inkscape or Adobe Illustrator.
pdf("outputs/plot2.pdf", width = 8, height = 6)
plot(x, y, main = "Scatter Plot", col = "blue") # first page
hist(x2, main = "Histogram")                    # second page
plot(x, y, main = "Line Plot", col = "red")     # third page
dev.off()
png 
  2

par function

  • par function is used to set large number of graphical parameters
  • It can be used to set the layout of multiple plots
    • par(mfrow = c(2, 2)) will create a 2x2 grid layout
  • par function can be used to set many graphical parameters like margins, font size, font family, etc.
    • par(mar = c(5, 4, 4, 2) + 0.1) will set the margins of the plot
  • par function can be used to set graphical parameters for text, lines, points, etc.
    • par(cex = 1.5) will set the size of text to 1.5 times the default size
  • par function can be used to set graphical parameters for colors, line types, line widths, etc.
    • par(lty = 2) will set the line type to dashed
  • see ?par for more details

Summary

  • Scatter plot: plot(x, y)
  • Line plot: plot(x, y, type = "l")
  • Box plot: boxplot(len ~ supp, data = ToothGrowth)
  • Bar plot: barplot(counts)
  • Histogram: hist(x)
  • Stripchart: stripchart(len ~ supp, data = ToothGrowth)
  • Exporting plots: png(), pdf()
  • Customizing plots: par()
  • Combining plots: par(mfrow = c(2, 2)), layout()